THE WARPED CIRCUMSTELLAR DISK of HD 100546 Alice C
Total Page:16
File Type:pdf, Size:1020Kb
The Astrophysical Journal, 640:1078–1085, 2006 April 1 # 2006. The American Astronomical Society. All rights reserved. Printed in U.S.A. THE WARPED CIRCUMSTELLAR DISK OF HD 100546 Alice C. Quillen Department of Physics and Astronomy, University of Rochester, Rochester, NY 14627; and Research School of Astronomy and Astrophysics, Australian National University, Mount Stromlo Observatory, Cotter Road, Weston Creek, ACT 2611, Australia; [email protected] Receivedv 2005 May 11; accepted 2005 November 30 ABSTRACT We propose that the two-armed spiral features seen in visible Hubble Space Telescope images of scattered light in HD 100546’s circumstellar disk are caused by the illumination of a warped outer disk. A tilt of 6 –15 from the symmetry plane can cause the observed surface brightness variations, providing the disk is very twisted (highly warped) at radii greater than 200 AU where the spiral features are seen. Dust lanes are due in part to shadowing in the equatorial plane from the inner disk within a radius of 100 AU. HD 100546’s outer disk, if viewed edge-on, would appear similar to that of Beta Pictoris. A disk initially misaligned with a planetary system becomes warped due to precession induced by planetesimal bodies and planets. However, the twistedness of HD 100546’s disk cannot be explained by precession during the lifetime of the system induced by a single Jovian-mass planet within the clearing at 13 AU. One possible explanation for the corrugated disk is that precession was induced by massive bodies embedded in the disk at larger radius. This would require approximately a Jupiter mass of bodies well outside the central clearing at 13 AU and within the location of the spiral features or at radii approximately between 50 and 200 AU. Subject headingg: stars: individual (HD 100546) 1. INTRODUCTION scenario extremely low, but it also implies that the perturbing object is near HD 100546. A search for a perturbing object (star) High angular resolution imaging of circumstellar disks have massive enough to account for the two arms was negative. shown that these disks may not be coplanar. For example, the disk Quillen et al. (2005) concluded that the spiral features were not of Beta Pictoris exhibits a warp; the outer disk at radii greater than density enhancements resulting from the tidal force of a recent 50 AU is tilted between 3 and 5 from the disk interior (Burrows stellar close encounter. In contrast, both Quillen et al. (2005) et al. 1995; Heap et al. 2000). Models for the scattered optical light and Ardila et al. (2005) have shown that the spiral arms in the of the bowed disk of AU Microscopii suggest that this disk is outer part of HD 141569A’s disk are likely to have been pulled also warped, with a small tilt of about 3 (Krist et al. 2005). While out tidally by the nearby passage of its binary companion, HD warped disks are easiest to identify in nearly edge-on systems 141569B and C. such as Beta Pictoris and AU Microscopii, less inclined or nearly Warped disks can display complex morphology. Previous face-on disks may also not be coplanar. This leads us to search work modeling galactic warped disks (e.g., Quillen et al. 1992, for features in less inclined disks that might also be explained 1993, 2006; Quillen & Bower 1999; Nicholson et al. 1992; with a warped disk model. As proposed by Mouillet et al. (1997) Tubbs 1980; Steiman-Cameron et al. 1992) accounts for the and Augereau et al. (2001b), the twistedness of Beta Pictoris’s optical dust lanes seen in Centaurus A and other galaxies and warped disk could be a result of precession induced by unseen the parallelogram-shaped infrared morphology seen in Spitzer planets or planetesimals residing in the disk. Consequently, by images of Centaurus A. These previous studies have shown that probing the geometry of the warp, we can constrain the prop- the apparent morphology of a warped disk can be sensitive to its erties of the unseen planetary system. optical properties (e.g., Krist et al. 2005) as well as to its ge- HD 100546 is a nearby southern Herbig Be star (B9.5 Ve; ometry. For emitting optically thin disks, folds in the disk can distance d ¼ 103 Æ 6 pc) with an estimated age t 10 Myr appear brighter to the viewer than other regions (see Quillen et al. (van den Ancker et al. 1997) that exhibits a dusty circumstellar [2006] and associated simulations). Likewise, we expect that op- disk revealed in scattered light from visible and near-infrared tically thin disks appear brighter in scattered light in regions imaging (Grady et al. 2001; Augereau et al. 2001a). We note that where the disk tilts at high inclination with respect to the viewer. at the distance of HD 100546, 100 corresponds approximately This is because the surface brightness of an optically thin medium to 100 AU. Most noticeable in the images of this disk taken by of uniform thickness that scatters light from a distant source is the Hubble Space Telescope (HST ) using the Space Telescope dependent on its orientation angle with respect to the viewer. We Imaging Spectrograph (STIS) camera are the two nearly sym- note that a disk that is optically thin when viewed from above metrical spiral features at a radius of about 250 AU from the star (integrated vertically) may be optically thick when viewed edge- (Grady et al. 2001). Quillen et al. (2005) showed that the two on. Folds in a warped disk could also be optically thick (as ex- arms were unlikely to be caused by spiral density waves excited plored in the case of Centaurus A in the near-infrared; Quillen by a planet embedded in the disk but could be density variations et al. 1993). pulled out tidally by a nearby passing star. However, Quillen Here we consider the possibility that the disk of HD 100546 et al. (2005) then ruled out this scenario for a number of reasons. is warped and that the warp is responsible for the observed To match the openness of the arms, the flyby had to be recent: spiral features. We first consider the case in which the outer disk within a few thousand years. Not only is the probability for this is isotropically illuminated by the central star. If the disk at large 1078 WARPED CIRCUMSTELLAR DISK OF HD 100546 1079 radii is warped and optically thick, then we would expect that model surface brightness images that can be compared to the the optical images would exhibit only one spiral arm. Only the observations. In x 3 we discuss dynamical or physical mod- high side of the disk would be illuminated, whereas the opposite els for the disk warp based on the geometric model that best side would be in shadow. If the disk is optically thin (even at matches the observed disk morphology. A summary and dis- folds) and warped at radii greater than 200 or 200 AU, then we cussion follows. expect to see two spiral features. Regions of high inclination with respect to the viewer would appear brighter and would be 2. WARPED DISK MODELS located on opposite sides of the star. We now consider the situation of a warped disk that is not 2.1. Warp Geometry isotropically illuminated. We expect that starlight illuminating We first discuss our notation for describing the orientation of the outer disk should be highly attenuated in the equatorial plane. the planetary system with respect to the viewer. The orientation Awarped disk can be described as a series of rings. At each radius of a coplanar planetary system requires two angles to describe: an inclined disk crosses the equatorial plane at two opposing , corresponding to the position angle (counterclockwise from points. Consequently, the shadow from the inner disk could cause north) of the axis of disk rotation on the sky, and an inclination two spiral dust lanes or shadows in the outer warped disk, re- angle, #, that describes the tilt of this axis with respect to the line sulting in the appearance of two spiral features. If the spiral fea- of sight. If # is zero, then the planetary system is viewed face-on. tures are due to an equatorial shadow, then the outer disk could For a system that is not coplanar, and # refer to the orientation be either optically thin or thick. If the outer disk is optically thick, of the rotation axis corresponding to the total angular momentum then one would expect two spiral shadows and one bright spiral of the system’s disk and planets. feature, whereas if the outer disk is optically thin, then one would We describe the warped disk with respect to the rotation axis expect two spiral shadows with brighter spiral features lying be- of the planetary system. A warped disk undergoing circular tween them. motion can be described as a series of rotating tilted rings, each Unfortunately, it is not straightforward to estimate the disk with a different radius, r. The orientation of each ring is specified normal optical depth and width of HD 100546’s disk as a func- by two angles, a precession angle, (r), similar to the longitude tion of radius. Based on the near-IR scattered light surface of the accenting node, and an inclination angle, i(r). These brightness, Pantin et al. (2000) estimated that the disk has a high angles are given with respect to the rotation axis of the plane- normal optical depth, 1, at radii greater than 100 or 100 AU. tary system and the direction of the line of sight.